Pith. sign in

REVIEW 1 major objections 4 minor 283 references

Test-Time Scaling in Reasoning LLMs: Inference Regimes, Evaluation, and Reproducibility

T0 review · 1 major / 4 minor · reviewed 2026-08-15 · deepseek-v4-flash

Pith's one-line read Test-time scaling results are properties of the full inference protocol, not of a scalar compute budget.

desk verdict Careful framework paper whose core argument holds, but the headline release count is wrong by three orders of magnitude and must be corrected before the corpus claims are credible. read the letter →

arxiv 2608.04001 v1 pith:GHTU5OQC submitted 2026-08-04 cs.LG cs.AI

classification cs.LGcs.AI
keywords test-timescalingreasoningLLMsinferenceprotocolsevaluationreproducibilitydiscovery-stabilityprofilerepeated-samplingmetricsprefix-treesearch
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

This paper argues that "test-time scaling" is not a single method but a family of budgeted inference algorithms over an autoregressive model's implicit prefix tree, and that measured performance depends on the entire inference system: checkpoint, prompt, decoder, search or aggregation controller, verifier, budget, and stopping rule. It distinguishes three structural regimes—single-trajectory sequential scaling, leaf-level sampling with terminal reduction, and prefix-level search—and provides an evaluation profile whose coordinates recover or bound common repeated-sampling metrics such as Pass@k and majority-vote consistency. If the paper is right, comparing reasoning systems requires reporting the full inference protocol together with any accuracy number, not just a scalar budget. The paper supports this with fixed response banks across broad knowledge, symbolic reasoning, and competition mathematics, and releases a large corpus of full reasoning traces for reuse.

What carries the argument

The load-bearing object is the implicit prefix tree $T(x)$ induced by an autoregressive model $p_\theta$ from prompt $x$, with terminal leaves $L(x)$; every test-time algorithm is viewed as a budgeted procedure $A_B$ that interleaves primitive operations (token generation, prefix expansion, verifier or judge calls, terminal reduction) under a cumulative cost bound $B$, yielding an induced proposal distribution $q_\pi(y|x)$ over completed leaves. The evaluation machinery is the discovery–stability profile $S_{k,t} = \frac{1}{Q}\sum_q \kappa_{k,t}(p_q)$, where $\kappa_{k,t}$ is the binomial tail probability that $k$ fresh attempts yield at least $t$ correct candidates; Pass@k, pass_k, majority-vote consistency, and related metrics are coordinates, functionals, or bounds of this profile. The framework also decomposes protocol cost as $C_{\mathrm{total}} = C_{\mathrm{gen}} + C_{\mathrm{eval}}$ with $C_{\mathrm{eval}} = C_{\mathrm{signal}} + C_{\mathrm{control}} + C_{\mathrm{decision}}$, and an energy–gate decomposition of token decoders defines the proposal distribution induced by truncation and reweighting rules.

What would settle it

For a fixed checkpoint and question set, generate two response banks that differ only in decoder gate order (top-p then top-k versus top-k then top-p) or only in the reducer, then compare accuracy across many prompts under protocol-matched bootstrap intervals; if the protocol changes move reported accuracy by less than the uncertainty in repeated runs, the central claim of protocol dependence is weakened, while systematic shifts beyond the intervals would confirm it.

Watch

Extended reading notes

Core claim

The paper's central claim is that test-time scaling should be formalized as a family of budgeted inference algorithms over the implicit prefix tree of an autoregressive model, with performance treated as a property of the complete inference system rather than of model weights or a single scalar budget. It distinguishes three regimes: single-trajectory sequential scaling, where one active path is extended or revised; leaf-level scaling, where a bank of completed candidates is reduced by voting, verifier selection, or minimum Bayes risk; and prefix-level scaling, where search over unfinished states uses continuation-value estimates and controllers such as beam, best-first, or MCTS-style search. The paper then develops evaluation principles: end-to-end system performance $M_B$ and candidate-bank diagnostics are different estimands; the discovery–stability profile $S_{k,t}$ unifies repeated-sampling metrics; and reproducibility is separated into exact replay versus distributional reproducibility, with uncertainty propagated through the actual aggregation and stopping protocol. Empirically, the paper shows on fixed response banks that candidate discovery outpaces selection: median Pass@80 reaches 82% while the all-correct coordinate falls to 15%, and swapping one reducer for another can move a configuration's accuracy from 75.6% down to 65.8% as the bank grows.

Load-bearing premise

The framework assumes that every test-time algorithm can be broken into primitive operations whose costs add up, and that the process gives a stable distribution over completed candidates; if sharing computation between prefixes or stop-when-you-want rules makes costs non-additive or makes the candidates dependent, the quantitative claims need to be reworked.

Editorial extensions

If this is right

  • Comparing reasoning models by a scalar budget and an accuracy number alone is insufficient; reports should specify the checkpoint, prompt, decoder, controller or reducer, verifier or judge, budget decomposition, and stopping rule together.
  • Common repeated-sampling metrics are not competing measures but coordinates or bounds of one discovery–stability profile, so they can be reported coherently and related to each other.
  • Shared-bank comparisons isolate the aggregation stage, end-to-end comparisons measure the deployed system, and compute accounting must include generation, evaluation, control, and decision costs rather than only sample counts.
  • Released response banks allow new reducers and verifiers to be evaluated without regenerating candidates, supporting exact replay of the paper's analyses.
  • Distributional reproducibility requires uncertainty estimates that propagate through the aggregation and stopping protocol, such as prompt bootstrap intervals computed after the reducer, rather than a single seeded run.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • This suggests a practical test: if every leaderboard entry carried a protocol fingerprint (prompt hash, decoder gate order, verifier identity, budget decomposition, stopping rule), many apparent ranking disagreements between studies would likely resolve into protocol differences rather than model differences.
  • The profile's discovery–stability shape also invites a two-axis performance comparison: plotting Pass@k against all-correct consistency would make the trade-off between finding a correct answer and finding it reliably the explicit object of comparison, rather than hiding it in a scalar.
  • Because the additive-cost premise is the framework's weakest point, a natural extension is to replace token-count budgets with wall-clock budgets under tree-structured inference kernels that share key–value cache; the paper notes such sharing can change the practical compute frontier but does not build it into the formal accounting.
Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

1 major / 4 minor

Summary. The paper develops a systematic account of test-time scaling for reasoning LLMs, formalizing it as budgeted inference over the implicit prefix tree of an autoregressive model and distinguishing single-trajectory sequential scaling, leaf-level scaling with terminal reduction, and prefix-level search. It proposes evaluation principles that treat the entire inference system as the evaluated object, introduces a discovery–stability profile that unifies repeated-sampling metrics, distinguishes exact replay from distributional reproducibility, and specifies protocol-matched reporting of compute and uncertainty. The framework is instantiated on MMLU-Pro, BBH, competition mathematics, and SuperGPQA with a large released corpus of reasoning traces, and the empirical sections include shared banks, bootstrap intervals, verifier diagnostics, and explicit caveats about token caps and effort confounding.

Significance. If the claims hold, the paper provides a genuinely useful vocabulary and evaluation protocol for test-time scaling: the three-regime taxonomy, the shared-bank versus end-to-end distinction, and the discovery–stability profile are well designed and fill a real gap in the literature. The empirical study is unusually careful: bootstrap intervals are propagated through the actual aggregation protocol, candidate banks are shared across reducers, token-cap rates and length limits are disclosed, verifier limitations (reference-assisted versus reference-free, ordinal rather than calibrated) are stated, and per-candidate seeds and serialization are specified. These practices are exactly the kind the paper argues for, and the release, once its size is stated correctly, will be a reusable resource. The central argument that reported performance depends on the full inference protocol is well supported by the formalization and the experiments.

major comments (1)
  1. [Abstract and Section 1 (release-size claim)] The Abstract states “assemble over 2 billion full reasoning traces” and the Introduction states “over 2 million reasoning traces,” but Table 1 and Section 5 report a total of 1,948,821 traces. The only 2-billion-scale number in the paper is the 1,832,922,842 “completion positions” in the compact Parquet snapshot described in Section G.5, which is a count of token positions, not full reasoning traces; the Introduction’s “over 2 million” also exceeds the tabulated total of 1,948,821. Because the release is a principal stated contribution and the paper itself (Section 3.5) argues that artifact counts are reproducibility requirements, this internal inconsistency is load-bearing. Please correct both statements (for example, “1,948,821 traces, including over 1.8 billion completion positions in the signal-rich snapshot”) and verify that every occurrence count in the abstract, introduction, body, and conclusion agrees with the release census.
minor comments (4)
  1. [Section 3.4 (Beta prior)] The default Beta(1,1) prior for the latent success probabilities p_q is a modeling choice; a short sensitivity check (for example, Beta(2,2) or an empirical-Bayes prior derived from an auxiliary bank) would strengthen the claim that the discovery–stability profile comparisons are not prior-driven.
  2. [Sections 2.1 and 3.3 (cost accounting)] The decomposition C_total = C_gen + C_eval assumes additive costs; since Section C.3 correctly notes that tree-structured kernels can reuse key–value caches and change the wall-clock frontier, the main text should state explicitly that the decomposition is a reporting convention rather than a physical identity, and that incommensurable costs should be reported separately.
  3. [Section G.3 (exact replay metadata)] For the broad knowledge and symbolic reasoning block, the dataset-hub revisions are unavailable and no chat template is recorded; these omissions are honestly disclosed, but given Section 3.5’s exact-replay requirements, please mark each release block as supporting exact replay, distributional reproducibility, or both, so readers know which artifacts are sufficient for which claim.
  4. [Section 4.3 and Table 14 (effort confounding)] The 16.12% token-cap rate for gpt-oss high effort is correctly reported, but the cap rate should appear alongside the main accuracy numbers in the main text rather than only in the appendix, since it directly qualifies the effort comparison and is a model of the disclosure the paper recommends.

Circularity Check

0 steps flagged · score 2.0 of 10

No material circularity: the framework is self-contained and the empirical claims are descriptive; only minor non-load-bearing self-citations appear.

full rationale

The paper's central claims are definitional and empirical, not derived from fitted parameters. The three-regime taxonomy and the prefix-tree formalization in Section 2.1 are explicit modeling choices; the evaluation profile in Section 3.4 is defined as a binomial-tail expectation, and the identities Pass@k=S_{k,1}, pass_k=S_{k,k}, and Maj@k≥S_{k,⌊k/2⌋+1} are algebraic consequences of that definition rather than predictions. The paper labels the profile a candidate-bank diagnostic and repeatedly states that it does not give the success probability of an aggregation rule, so no fitted value is renamed as a result. The empirical study uses fixed response banks and exact without-replacement statistics, with no parameter fitted to benchmark outcomes and then called a prediction. The only self-citations (Hariri et al. 2026a,c) support tie-breaking conventions, ranking uncertainty, and reproducibility framing; the present paper re-derives its framework in full, so these citations are not load-bearing. Acknowledged limitations, including ordinal verifier calibration, the reference-assisted Compass diagnostic, search-induced candidate dependence, and the high-effort token-cap confounding, are honest disclosures rather than circular steps. One correctness concern outside circularity remains: the abstract and contribution list claim 'over 2 billion full reasoning traces' while Table 1 and Section 5 total 1,948,821 traces, and the only 1.8-billion figure in Section G.5 counts completion token positions; this is an internal release-statistic inconsistency, not a definitional reduction.

Assumptions & free parameters 3 free parameters · 5 assumptions · 1 invented entities

The central framework rests on domain assumptions about additivity of compute costs and well-defined proposal distributions; the empirical diagnostics rest on Bernoulli/Beta modeling for candidate banks. No physical entities are posited; the only new conceptual object is the discovery-stability profile, which is a measurement construct rather than a falsifiable prediction.

free parameters (3)
  • Beta(1,1) prior for latent prompt success probabilities p_q
    The Bayesian profile report defaults to a uniform Beta(1,1) prior per prompt. It is a modeling choice, not fitted, but the posterior summaries depend on it.
  • Contextual Compass null-adjustment temperature = 1.5
    The contextual CompassVerifier score divides the difference between slot and null log-probabilities by 1.5, a hand-selected constant for this diagnostic.
  • Pointwise Qwen ordinal score mapping = A=20 ... T=1
    The reference-free verifier maps ordinal A-T to 20..1 and normalizes to [0,1], a hand-chosen scoring convention used only in the empirical diagnostics.
assumptions (5)
  • domain assumption An autoregressive language model and a local generation policy induce a well-defined proposal distribution over completed leaves via the implicit prefix tree.
    Section 2.1 builds the entire formalization on this prefix-tree model, which is standard for token-level LLM decoding.
  • domain assumption Task utility U_x(y) is a well-defined scalar for benchmarks, either exact correctness or a graded score.
    The end-to-end estimand and profile definitions in Sections 2.1 and 3.4 assume a well-defined utility; the paper notes open-ended tasks require the evaluation protocol to be part of the method.
  • domain assumption Compute costs are additive and can be split into generation, signal, control, and decision costs.
    The compute accounting C_total = C_gen + C_eval in Section 3.3 relies on this; the paper itself notes that incommensurate costs should be reported separately.
  • domain assumption For the independent-attempt profile, latent per-prompt success probabilities p_q exist and the Beta-Binomial model applies.
    The latent discovery-stability profile and posterior mean formulas in Section 3.4 require this; the paper states the i.i.d. interpretation should only be used when it matches the sampling protocol.
  • standard math Standard probability identities are accepted: binomial tails, hypergeometric without-replacement estimates, beta-binomial conjugacy, and the MBR equivalence of self-consistency.
    Sections 3.4 and Appendix B use these without proof, which is appropriate for standard results.
invented entities (1)
  • Discovery-stability profile S_k
    purpose: Summarize a candidate bank by, for each threshold t, the expected fraction of prompts with at least t correct candidates among k attempts; it unifies Pass@k, pass@k, Maj@k bounds, and other repeated-sampling metrics.
    This is a new statistical summary object introduced by the paper, not a physical entity. It is a measurement construct rather than a falsifiable prediction, so it has no external empirical handle.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Test-Time Scaling in Reasoning LLMs: Inference Regimes, Evaluation, and Reproducibility." pith.science (2026). https://pith.science/paper/GHTU5OQC

@misc{pith2026260804001,
  author       = {Pith},
  title        = {Pith review of: Test-Time Scaling in Reasoning LLMs: Inference Regimes, Evaluation, and Reproducibility},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/GHTU5OQC}},
  note         = {Machine review of arXiv:2608.04001}
}
read the original abstract

Large language models can solve substantially harder reasoning problems with more inference-time compute. The term "test-time scaling," however, now covers diverse inference algorithms that extend deliberation along a single trajectory, sample completed candidates and aggregate them through voting or verification, or search over unfinished partial states. These algorithms differ in their statistical structure, compute accounting, and failure modes. Treating these procedures as interchangeable under a single scalar "budget," or reporting accuracy without the inference protocol that produced it, makes results difficult to compare across studies. We develop a systematic account of test-time scaling along three axes. First, we formalize test-time scaling as budgeted inference over the implicit prefix tree of an autoregressive model and distinguish three structural regimes: single-trajectory sequential scaling, leaf-level scaling with terminal reduction, and prefix-level scaling. Second, we treat the evaluated object as the entire inference system and develop evaluation principles that separate end-to-end system performance from candidate-bank diagnostics. We introduce an evaluation profile whose coordinates and simple functionals recover or bound common repeated-sampling metrics, and prescribe protocol-matched reporting of compute and uncertainty. Third, we specify reproducibility requirements for inference protocols, distinguishing exact replay from distributional reproducibility and identifying the artifacts needed to support each. We also organize the open-weight reasoning ecosystem by model-side and interface mechanisms, apply these principles to broad-knowledge, symbolic-reasoning, and competition-mathematics benchmarks, and assemble over 2 billion full reasoning traces for release with progressively richer verifier and token-level signals.

Figures

Figures reproduced from arXiv: 2608.04001 by the authors.

Figure 1
Figure 1. A selective chronicle of reasoning systems, from elicited chains to budgeted inference. Test-time in [PITH_FULL_IMAGE:figures/full_fig_p003_1.png] view at source ↗
Figure 2
Figure 2. Three regimes of test-time scaling over the implicit prefix tree. Left: single-trajectory sequential [PITH_FULL_IMAGE:figures/full_fig_p005_2.png] view at source ↗
Figure 3
Figure 3. Elementwise view of a one-token sampling protocol. The sequence-level proposal distribution [PITH_FULL_IMAGE:figures/full_fig_p011_3.png] view at source ↗
Figures from the paper (5 more)
Figure 4
Figure 4. Figure 4: Aggregation and stopping under sampled inference. Evidence extraction and fixed-bank selection [PITH_FULL_IMAGE:figures/full_fig_p013_4.png]
Figure 5
Figure 5. Figure 5: Candidate availability and reducer accuracy under repeated sampling. (a) Median exact Pass@k and passk across the 20 configurations; bands span the interquartile range of configurations. (b) Mean agreement of subset rankings with the N = 80 ranking; bands are central 9…
Figure 6
Figure 6. Figure 6: Finite-bank scaling on the five 2025–2026 competition sets. Each panel compares the exact Pass@k ceiling with literal answer plurality and pointwise Best-of-N using the reference-free verifier described below. Arrows mark the k = 80 gap between Pass@80 and the better o…
Figure 7
Figure 7. Figure 7: Verifier signals against the independent rule-based outcome on 104,160 competition￾math traces. The figure compares direct and contextual Compass A scores with the reference-free point￾wise Qwen score through conditional distributions and equal-count score bins. The lo…
Figure 8
Figure 8. Figure 8: SuperGPQA scaling and field heterogeneity for the finalized gpt-oss high bank. The left panel gives exact finite-bank discovery and stability over 3,600 questions. In the right panel, each point is one of 72 equally sized fields: the axes compare single-response accura…

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

283 extracted references · 16 canonical work pages

  1. [1]

    arXiv preprint arXiv:1904.09751 , year=

    The curious case of neural text degeneration , author=. arXiv preprint arXiv:1904.09751 , year=

  2. [2]

    arXiv preprint arXiv:2408.10577 , year=

    Optimizing large language model hyperparameters for code generation , author=. arXiv preprint arXiv:2408.10577 , year=

  3. [3]

    arXiv preprint arXiv:2407.01082 , year=

    Turning up the heat: Min-p sampling for creative and coherent llm outputs , author=. arXiv preprint arXiv:2407.01082 , year=

  4. [4]

    arXiv preprint arXiv:2506.13681 , year=

    Turning Down the Heat: A Critical Analysis of Min-p Sampling in Language Models , author=. arXiv preprint arXiv:2506.13681 , year=

  5. [5]

    arXiv preprint arXiv:2402.06925 , year=

    A thorough examination of decoding methods in the era of llms , author=. arXiv preprint arXiv:2402.06925 , year=

  6. [6]

    arXiv preprint arXiv:2310.01693 , year=

    Closing the curious case of neural text degeneration , author=. arXiv preprint arXiv:2310.01693 , year=

  7. [8]

    arXiv preprint arXiv:2407.21787 , year=

    Large language monkeys: Scaling inference compute with repeated sampling , author=. arXiv preprint arXiv:2407.21787 , year=

  8. [9]

    Proceedings of the 2025 Conference on Empirical Methods in Natural Language Processing , pages =

    s1: Simple Test-Time Scaling , author =. Proceedings of the 2025 Conference on Empirical Methods in Natural Language Processing , pages =. 2025 , publisher =

Show all 283 references
  1. [10]

    B leu: a Method for Automatic Evaluation of Machine Translation

    Papineni, Kishore and Roukos, Salim and Ward, Todd and Zhu, Wei-Jing. B leu: a Method for Automatic Evaluation of Machine Translation. Proceedings of the 40th Annual Meeting of the Association for Computational Linguistics. 2002. doi:10.3115/1073083.1073135

  2. [11]

    arXiv preprint arXiv:2009.10297 , year=

    Codebleu: a method for automatic evaluation of code synthesis , author=. arXiv preprint arXiv:2009.10297 , year=

  3. [12]

    Advances in Neural Information Processing Systems , volume=

    Spoc: Search-based pseudocode to code , author=. Advances in Neural Information Processing Systems , volume=

  4. [15]

    arXiv preprint arXiv:2203.11171 , year=

    Self-consistency improves chain of thought reasoning in language models , author=. arXiv preprint arXiv:2203.11171 , year=

  5. [16]

    Advances in neural information processing systems , volume=

    Solving quantitative reasoning problems with language models , author=. Advances in neural information processing systems , volume=

  6. [17]

    2024 , eprint=

    Yao, Shunyu and Shinn, Noah and Razavi, Pedram and Narasimhan, Karthik , journal=. 2024 , eprint=. doi:10.48550/arXiv.2406.12045 , note=

  7. [18]

    arXiv preprint arXiv:2411.00640 , year=

    Adding error bars to evals: A statistical approach to language model evaluations , author=. arXiv preprint arXiv:2411.00640 , year=

  8. [19]

    arXiv preprint arXiv:2504.21303 , year=

    Confidence in large language model evaluation: A bayesian approach to limited-sample challenges , author=. arXiv preprint arXiv:2504.21303 , year=

  9. [20]

    Advances in Neural Information Processing Systems , volume=

    Benchmarking llms via uncertainty quantification , author=. Advances in Neural Information Processing Systems , volume=

  10. [21]

    The Fourteenth International Conference on Learning Representations , year=

    Don't Pass@k: A Bayesian Framework for Large Language Model Evaluation , author=. The Fourteenth International Conference on Learning Representations , year=

  11. [23]

    The Annals of Statistics , volume=

    Bootstrap Methods: Another Look at the Jackknife , author=. The Annals of Statistics , volume=

  12. [24]

    Forty-first International Conference on Machine Learning , year=

    Chatbot arena: An open platform for evaluating llms by human preference , author=. Forty-first International Conference on Machine Learning , year=

  13. [25]

    arXiv preprint arXiv:2412.18407 , year=

    A statistical framework for ranking llm-based chatbots , author=. arXiv preprint arXiv:2412.18407 , year=

  14. [26]

    arXiv preprint arXiv:2508.11452 , year=

    Inclusion Arena: An Open Platform for Evaluating Large Foundation Models with Real-World Apps , author=. arXiv preprint arXiv:2508.11452 , year=

  15. [27]

    2nd Workshop on Models of Human Feedback for AI Alignment , year=

    Dropping Just a Handful of Preferences Can Change Top Large Language Model Rankings , author=. 2nd Workshop on Models of Human Feedback for AI Alignment , year=

  16. [29]

    Findings of the Association for Computational Linguistics: NAACL 2025 , pages=

    Re-evaluating automatic LLM system ranking for alignment with human preference , author=. Findings of the Association for Computational Linguistics: NAACL 2025 , pages=

  17. [30]

    Improving Reproducibility in Machine Learning Research (A Report from the

    Pineau, Joelle and Vincent-Lamarre, Philippe and Sinha, Koustuv and Larivi. Improving Reproducibility in Machine Learning Research (A Report from the. Journal of Machine Learning Research , volume =. 2021 , url =

  18. [32]

    Proceedings of Machine Learning and Systems , volume =

    Accounting for Variance in Machine Learning Benchmarks , author =. Proceedings of Machine Learning and Systems , volume =. 2021 , url =

  19. [34]

    Second Conference on Language Modeling , year =

    A Sober Look at Progress in Language Model Reasoning: Pitfalls and Paths to Reproducibility , author =. Second Conference on Language Modeling , year =

  20. [35]

    Yuan, Jiayi and Li, Hao and Ding, Xinheng and Xie, Wenya and Li, Yu-Jhe and Zhao, Wentian and Wan, Kun and Shi, Jing and Hu, Xia and Liu, Zirui , journal =. Give Me. 2025 , url =

  21. [36]

    arXiv preprint arXiv:2504.04823 , year=

    Quantization hurts reasoning? an empirical study on quantized reasoning models , author=. arXiv preprint arXiv:2504.04823 , year=

  22. [38]

    , booktitle=

    Tahmasivand, Ahmad and Zahran, Noureldin and Al-Sayouri, Saba and Fouda, Mohammed and Khasawneh, Khaled N. , booktitle=. LM-Fix: Lightweight Bit-Flip Detection and Rapid Recovery Framework for Language Models , year=

  23. [39]

    2025 , archivePrefix=

    INTERLACE: Interleaved Layer Pruning and Efficient Adaptation in Large Vision-Language Models , author=. 2025 , archivePrefix=

  24. [40]

    Transactions on Machine Learning Research , year=

    Holistic evaluation of language models , author=. Transactions on Machine Learning Research , year=

  25. [41]

    Proceedings of the 2021 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies , pages=

    Dynabench: Rethinking benchmarking in NLP , author=. Proceedings of the 2021 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies , pages=. 2021 , url=

  26. [42]

    International Conference on Learning Representations , year=

    Proving test set contamination in black-box language models , author=. International Conference on Learning Representations , year=

  27. [43]

    Nature , volume=

    A benchmark of expert-level academic questions to assess AI capabilities , author=. Nature , volume=. 2026 , doi=

  28. [44]

    International Conference on Machine Learning , year=

    DyCodeEval: Dynamic Benchmarking of Reasoning Capabilities in Code Large Language Models Under Data Contamination , author=. International Conference on Machine Learning , year=

  29. [45]

    International Conference on Learning Representations , year=

    LiveBench: A Challenging, Contamination-Limited LLM Benchmark , author=. International Conference on Learning Representations , year=

  30. [46]

    First Conference on Language Modeling , year=

    Gpqa: A graduate-level google-proof q&a benchmark , author=. First Conference on Language Modeling , year=

  31. [47]

    arXiv preprint arXiv:2311.07911 , year=

    Instruction-following evaluation for large language models , author=. arXiv preprint arXiv:2311.07911 , year=

  32. [48]

    Advances in Neural Information Processing Systems , volume =

    Wang, Yubo and Ma, Xueguang and Zhang, Ge and Ni, Yuansheng and Chandra, Abhranil and Guo, Shiguang and Ren, Weiming and Arulraj, Aaran and He, Xuan and Jiang, Ziyan and Li, Tianle and Ku, Max and Wang, Kai and Zhuang, Alex and Fan, Rongqi and Yue, Xiang and Chen, Wenhu , titl...

  33. [49]

    arXiv preprint arXiv:2502.06453 , year=

    MATH-Perturb: Benchmarking LLMs' Math Reasoning Abilities against Hard Perturbations , author=. arXiv preprint arXiv:2502.06453 , year=. doi:10.48550/arXiv.2502.06453 , url=

  34. [50]

    International Conference on Learning Representations , year=

    BeyondBench: Contamination-Resistant Evaluation of Reasoning in Language Models , author=. International Conference on Learning Representations , year=

  35. [51]

    arXiv preprint arXiv:2602.10367 , year=

    LiveMedBench: A Contamination-Free Medical Benchmark for LLMs with Automated Rubric Evaluation , author=. arXiv preprint arXiv:2602.10367 , year=. doi:10.48550/arXiv.2602.10367 , url=

  36. [52]

    Challenging

    Suzgun, Mirac and Scales, Nathan and Sch. Challenging. Findings of the Association for Computational Linguistics: ACL 2023 , pages =. 2023 , doi =

  37. [53]

    arXiv preprint arXiv:2103.03874 , year=

    Measuring mathematical problem solving with the math dataset , author=. arXiv preprint arXiv:2103.03874 , year=

  38. [54]

    2025 , archivePrefix=

    Simulating a Bias Mitigation Scenario in Large Language Models , author=. 2025 , archivePrefix=

  39. [55]

    2026 , archivePrefix=

    Understanding Users' Privacy Reasoning and Behaviors During Chatbot Use to Support Meaningful Agency in Privacy , author=. 2026 , archivePrefix=

  40. [56]

    2026 , doi =

    Sanitizing manufacturing dataset labels using vision-language models , journal =. 2026 , doi =

  41. [58]

    QwQ-32B: Embracing the Power of Reinforcement Learning , url =

  42. [59]

    arXiv preprint arXiv:2412.15115 , year=

    Qwen2.5 Technical Report , author=. arXiv preprint arXiv:2412.15115 , year=

  43. [60]

    Sky-T1: Train your own O1 preview model within \ 450 , howpublished =

  44. [61]

    Think Less, Achieve More: Cut Reasoning Costs by 50\ howpublished =

  45. [62]

    Unlocking the Potential of Reinforcement Learning in Improving Reasoning Models , howpublished =

  46. [63]

    arXiv preprint arXiv:2502.14382 , year=

    S*: Test Time Scaling for Code Generation , author=. arXiv preprint arXiv:2502.14382 , year=

  47. [64]

    arXiv preprint arXiv:2305.11206 , year=

    LIMA: Less Is More for Alignment , author=. arXiv preprint arXiv:2305.11206 , year=

  48. [66]

    arXiv preprint arXiv:2502.11886 , year=

    Limr: Less is more for rl scaling , author=. arXiv preprint arXiv:2502.11886 , year=

  49. [67]

    Bespoke-Stratos: The unreasonable effectiveness of reasoning distillation , howpublished =

  50. [68]

    Guha, Etash and Raoof, Negin and Mercat, Jean and Frankel, Eric and Keh, Sedrick and Grover, Sachin and Smyrnis, George and Vu, Trung and Marten, Ryan and Saad-Falcon, Jon and Choi, Caroline and Arora, Kushal and Merrill, Mike and Deng, Yichuan and Suvarna, Ashima and Bansal, ...

  51. [69]

    Open R1: A fully open reproduction of DeepSeek-R1 , url =

  52. [70]

    arXiv preprint arXiv:2401.10491 , year=

    Knowledge fusion of large language models , author=. arXiv preprint arXiv:2401.10491 , year=

  53. [71]

    arXiv preprint arXiv:2408.07990 , year=

    Fusechat: Knowledge fusion of chat models , author=. arXiv preprint arXiv:2408.07990 , year=

  54. [72]

    arXiv preprint arXiv:2503.04222 , year=

    Fusechat-3.0: Preference optimization meets heterogeneous model fusion , author=. arXiv preprint arXiv:2503.04222 , year=

  55. [73]

    arXiv preprint arXiv:2503.10460 , year=

    Light-r1: Curriculum sft, dpo and rl for long cot from scratch and beyond , author=. arXiv preprint arXiv:2503.10460 , year=

  56. [74]

    arXiv preprint arXiv:2503.04872 , year=

    Tinyr1-32b-preview: Boosting accuracy with branch-merge distillation , author=. arXiv preprint arXiv:2503.04872 , year=

  57. [76]

    The Twelfth International Conference on Learning Representations , year =

    Escape Sky-high Cost: Early-stopping Self-Consistency for Multi-step Reasoning , author =. The Twelfth International Conference on Learning Representations , year =

  58. [77]

    The Fourteenth International Conference on Learning Representations , year =

    Deep Think with Confidence , author =. The Fourteenth International Conference on Learning Representations , year =

  59. [78]

    Advances in Neural Information Processing Systems , volume =

    Scalable Best-of- N Selection for Large Language Models via Self-Certainty , author =. Advances in Neural Information Processing Systems , volume =. 2025 , url =

  60. [79]

    Optimal Aggregation of

    Kuang, Peng and Wang, Yanli and Han, Xiaoyu and Liu, Yaowenqi and Xu, Kaidi and Wang, Haohan , booktitle =. Optimal Aggregation of. 2026 , url =

  61. [80]

    Advances in Neural Information Processing Systems , volume =

    Majority of the Bests: Improving Best-of- N via Bootstrapping , author =. Advances in Neural Information Processing Systems , volume =. 2025 , url =

  62. [81]

    The Fourteenth International Conference on Learning Representations , year =

    Best-of-Majority: Minimax-Optimal Strategy for Pass@ k Inference Scaling , author =. The Fourteenth International Conference on Learning Representations , year =

  63. [82]

    Transactions on Machine Learning Research , year =

    From Decoding to Meta-Generation: Inference-time Algorithms for Large Language Models , author =. Transactions on Machine Learning Research , year =

  64. [83]

    Proceedings of the 40th International Conference on Machine Learning (ICML) , series =

    Scaling Laws for Reward Model Overoptimization , author =. Proceedings of the 40th International Conference on Machine Learning (ICML) , series =. 2023 , url =

  65. [84]

    Proceedings of the 42nd International Conference on Machine Learning (ICML) , series =

    Is Best-of- N the Best of Them? Coverage, Scaling, and Optimality in Inference-Time Alignment , author =. Proceedings of the 42nd International Conference on Machine Learning (ICML) , series =. 2025 , url =

  66. [86]

    2025 , eprint =

    Soft Best-of- n Sampling for Model Alignment , author =. 2025 , eprint =. doi:10.48550/arXiv.2505.03156 , note =

  67. [87]

    , year =

    Bertsch, Amanda and Xie, Alex and Neubig, Graham and Gormley, Matthew R. , year =. It's. doi:10.48550/arXiv.2310.01387 , url =. 2310.01387 , archivePrefix=

  68. [88]

    Proceedings of the Human Language Technology Conference of the North American Chapter of the Association for Computational Linguistics (HLT-NAACL) , year =

    Minimum Bayes-Risk Decoding for Statistical Machine Translation , author =. Proceedings of the Human Language Technology Conference of the North American Chapter of the Association for Computational Linguistics (HLT-NAACL) , year =

  69. [89]

    Transactions of the Association for Computational Linguistics , volume =

    High Quality Rather than High Model Probability: Minimum Bayes Risk Decoding with Neural Metrics , author =. Transactions of the Association for Computational Linguistics , volume =. 2022 , doi =

  70. [90]

    International Conference on Learning Representations (ICLR) , year =

    Better Instruction-Following Through Minimum Bayes Risk , author =. International Conference on Learning Representations (ICLR) , year =. doi:10.48550/arXiv.2410.02902 , url =. 2410.02902 , archivePrefix=

  71. [91]

    2023 , eprint =

    Faster Minimum Bayes Risk Decoding with Confidence-based Pruning , author =. 2023 , eprint =. doi:10.48550/arXiv.2311.14919 , url =

  72. [92]

    2024 , eprint =

    Hyperparameter-Free Approach for Faster Minimum Bayes Risk Decoding , author =. 2024 , eprint =. doi:10.48550/arXiv.2401.02749 , url =

  73. [93]

    Eikema, Bryan and Aziz, Wilker , booktitle =. Is. 2020 , url =

  74. [94]

    2022 , eprint =

    Self-Consistency Improves Chain of Thought Reasoning in Language Models , author =. 2022 , eprint =. doi:10.48550/arXiv.2203.11171 , url =

  75. [95]

    2023 , eprint =

    Universal Self-Consistency for Large Language Model Generation , author =. 2023 , eprint =. doi:10.48550/arXiv.2311.17311 , url =

  76. [96]

    2025 , eprint =

    Ranked Voting based Self-Consistency of Large Language Models , author =. 2025 , eprint =. doi:10.48550/arXiv.2505.10772 , note =

  77. [97]

    2021 , eprint =

    Training Verifiers to Solve Math Word Problems , author =. 2021 , eprint =. doi:10.48550/arXiv.2110.14168 , url =

  78. [98]

    Lightman, Hunter and others , year =. Let. doi:10.48550/arXiv.2305.20050 , url =. 2305.20050 , archivePrefix=

  79. [99]

    2024 , eprint =

    BoNBoN Alignment for Large Language Models and the Sweetness of Best-of- n Sampling , author =. 2024 , eprint =. doi:10.48550/arXiv.2406.00832 , url =

  80. [100]

    2025 , eprint =

    Variational Best-of- N Alignment , author =. 2025 , eprint =. doi:10.48550/arXiv.2407.06057 , note =

  81. [102]

    2023 , eprint =

    Tree of Thoughts: Deliberate Problem Solving with Large Language Models , author =. 2023 , eprint =. doi:10.48550/arXiv.2305.10601 , url =

  82. [103]

    2024 , month = jul, publisher =

    Wan, Ziyu and Feng, Xidong and Wen, Muning and Mcaleer, Stephen Marcus and Wen, Ying and Zhang, Weinan and Wang, Jun , booktitle =. 2024 , month = jul, publisher =

  83. [104]

    2024 , month = jun, publisher =

    Yu, Fei and Gao, Anningzhe and Wang, Benyou , booktitle =. 2024 , month = jun, publisher =. doi:10.18653/v1/2024.findings-naacl.55 , url =

  84. [105]

    Policy Guided Tree Search for Enhanced

    Li, Yang , booktitle =. Policy Guided Tree Search for Enhanced. 2025 , month = jul, publisher =

  85. [106]

    Transactions on Machine Learning Research , year =

    Tree Search for Language Model Agents , author =. Transactions on Machine Learning Research , year =

  86. [107]

    2026 , eprint =

    Policy-Guided Search on Tree-of-Thoughts for Efficient Problem Solving with Bounded Language Model Queries , author =. 2026 , eprint =. doi:10.48550/arXiv.2601.03606 , url =

  87. [108]

    2026 , note =

    On the Scaling Flaws of Verifier-Guided Beam Search , author =. 2026 , note =

  88. [109]

    Advances in Neural Information Processing Systems , year =

    Reasoning Is Not a Race: When Stopping Early Beats Going Deeper , author =. Advances in Neural Information Processing Systems , year =

  89. [110]

    DEFT: Decoding with Flash Tree-Attention for Efficient Tree-Structured

    Yao, Jinwei and Chen, Kaiqi and Zhang, Kexun and You, Jiaxuan and Yuan, Binhang and Wang, Zeke and Lin, Tao , booktitle =. DEFT: Decoding with Flash Tree-Attention for Efficient Tree-Structured. 2025 , url =

  90. [111]

    Advances in Neural Information Processing Systems , year =

    Thought of Search: Planning with Language Models Through The Lens of Efficiency , author =. Advances in Neural Information Processing Systems , year =

  91. [112]

    Bandit Based

    Kocsis, Levente and Szepesv. Bandit Based. Proceedings of the 17th European Conference on Machine Learning (ECML) , year =

  92. [113]

    and Powley, Edward and Whitehouse, Daniel and Lucas, Simon M

    Browne, Cameron B. and Powley, Edward and Whitehouse, Daniel and Lucas, Simon M. and Cowling, Peter I. and Rohlfshagen, Philipp and Tavener, Stephen and Perez, Diego and Samothrakis, Spyridon and Colton, Simon , journal =. A Survey of. 2012 , doi =

  93. [114]

    Proceedings of the 2025 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies (Volume 1: Long Papers) , pages =

    Reward-Guided Tree Search for Inference Time Alignment of Large Language Models , author =. Proceedings of the 2025 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies (Volume 1: Long Papers) , pages =. 2025 , url =

  94. [115]

    2024 , urldate =

    American Invitational Mathematics Examination (AIME) , author =. 2024 , urldate =

  95. [116]

    2025 , urldate =

    American Invitational Mathematics Examination (AIME) , author =. 2025 , urldate =

  96. [117]

    2025 , urldate =

    HMMT February 2025 Archive (Problems and Solutions) , author =. 2025 , urldate =

  97. [118]

    2025 , urldate =

    Brown University Math Olympiad (BrUMO) , author =. 2025 , urldate =

  98. [119]

    NovaSky Team , title =

  99. [120]

    2025 , eprint=

    Qwen3 Technical Report , author=. 2025 , eprint=

  100. [121]

    2025 , eprint=

    gpt-oss-120b & gpt-oss-20b Model Card , author=. 2025 , eprint=

  101. [122]

    2502.03387 , archivePrefix=

    Yixin Ye and Zhen Huang and Yang Xiao and Ethan Chern and Shijie Xia and Pengfei Liu , year=. 2502.03387 , archivePrefix=

  102. [123]

    arXiv preprint arXiv:2507.11407 , year=

    EXAONE 4.0: Unified Large Language Models Integrating Non-reasoning and Reasoning Modes , author=. arXiv preprint arXiv:2507.11407 , year=

  103. [124]

    2025 , eprint=

    Wasi Uddin Ahmad and Sean Narenthiran and Somshubra Majumdar and Aleksander Ficek and Siddhartha Jain and Jocelyn Huang and Vahid Noroozi and Boris Ginsburg , journal=. 2025 , eprint=

  104. [126]

    2504.16891 , archivePrefix=

    Ivan Moshkov and Darragh Hanley and Ivan Sorokin and Shubham Toshniwal and Christof Henkel and Benedikt Schifferer and Wei Du and Igor Gitman , year=. 2504.16891 , archivePrefix=

  105. [127]

    2025 , url=

    Shubham Toshniwal and Ivan Sorokin and Aleksander Ficek and Ivan Moshkov and Igor Gitman , booktitle=. 2025 , url=

  106. [128]

    and Zhang, Songyang and Chen, Kai , title =

    Liu, Shudong and Liu, Hongwei and Liu, Junnan and Xiao, Linchen and Gao, Songyang and Lyu, Chengqi and Gu, Yuzhe and Zhang, Wenwei and Wong, Derek F. and Zhang, Songyang and Chen, Kai , title =. Proceedings of the 2025 Conference on Empirical Methods in Natural Language Proces...

  107. [130]

    Proceedings of the 2025 Conference on Empirical Methods in Natural Language Processing , pages=

    Fusechat: Knowledge fusion of chat models , author=. Proceedings of the 2025 Conference on Empirical Methods in Natural Language Processing , pages=

  108. [131]

    Proceedings of the 63rd Annual Meeting of the Association for Computational Linguistics (Volume 6: Industry Track) , pages=

    Light-r1: Curriculum sft, dpo and rl for long cot from scratch and beyond , author=. Proceedings of the 63rd Annual Meeting of the Association for Computational Linguistics (Volume 6: Industry Track) , pages=

  109. [132]

    arXiv preprint arXiv:2506.13284 , year=

    AceReason-Nemotron 1.1: Advancing Math and Code Reasoning through SFT and RL Synergy , author=. arXiv preprint arXiv:2506.13284 , year=

  110. [134]

    Advances in Neural Information Processing Systems , year =

    70\ author =. Advances in Neural Information Processing Systems , year =

  111. [135]

    arXiv preprint arXiv:2307.08691 , year=

    Flashattention-2: Faster attention with better parallelism and work partitioning , author=. arXiv preprint arXiv:2307.08691 , year=

  112. [136]

    , author =

    Accelerate: Training and inference at scale made simple, efficient and adaptable. , author =

  113. [137]

    arXiv preprint arXiv:1910.03771 , year=

    Huggingface's transformers: State-of-the-art natural language processing , author=. arXiv preprint arXiv:1910.03771 , year=

  114. [138]

    and Zhang, Hao and Stoica, Ion , title =

    Kwon, Woosuk and Li, Zhuohan and Zhuang, Siyuan and Sheng, Ying and Zheng, Lianmin and Yu, Cody Hao and Gonzalez, Joseph E. and Zhang, Hao and Stoica, Ion , title =. Proceedings of the 29th Symposium on Operating Systems Principles , pages =. 2023 , publisher =. doi:10.1145/36...

  115. [139]

    Proceedings of the 14th International Conference on Learning Representations (ICLR 2026) , year =

    Don't Pass@k: A Bayesian Framework for Large Language Model Evaluation , author =. Proceedings of the 14th International Conference on Learning Representations (ICLR 2026) , year =. 2510.04265 , archivePrefix =

  116. [140]

    Competition-level code generation with

    Li, Yujia and Choi, David and others , journal=. Competition-level code generation with. 2022 , doi=

  117. [141]

    arXiv preprint arXiv:2408.00724 , year=

    Inference Scaling Laws: An Empirical Analysis of Compute-Optimal Inference for Problem-Solving with Language Models , author=. arXiv preprint arXiv:2408.00724 , year=

  118. [142]

    Snell, Charlie and Lee, Jaehoon and Xu, Kelvin and Kumar, Aviral , booktitle=. Scaling

  119. [143]

    Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing (EMNLP) , year=

    Improving Minimum Bayes Risk Decoding with Multi-Prompt , author=. Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing (EMNLP) , year=

  120. [144]

    arXiv preprint arXiv:2310.02743 , year=

    Reward Model Ensembles Help Mitigate Overoptimization , author=. arXiv preprint arXiv:2310.02743 , year=

  121. [145]

    Proceedings of the 63rd Annual Meeting of the Association for Computational Linguistics (ACL) , pages=

    Revisiting the Test-Time Scaling of o1-like Models: Do they Truly Possess Test-Time Scaling Capabilities? , author=. Proceedings of the 63rd Annual Meeting of the Association for Computational Linguistics (ACL) , pages=

  122. [146]

    arXiv preprint arXiv:2512.02008 , year=

    The Art of Scaling Test-Time Compute for Large Language Models , author=. arXiv preprint arXiv:2512.02008 , year=

  123. [147]

    Confidence Improves Self-Consistency in

    Taubenfeld, Amir and Sheffer, Tom and Ofek, Eran and Feder, Amir and Goldstein, Ariel and Gekhman, Zorik and Yona, Gal , booktitle =. Confidence Improves Self-Consistency in. 2025 , month = jul, publisher =

  124. [148]

    2024 , eprint =

    Semantic Self-Consistency: Enhancing Language Model Reasoning via Semantic Weighting , author =. 2024 , eprint =

  125. [149]

    and Zhang, Hao and Gonzalez, Joseph E

    Zheng, Lianmin and Chiang, Wei-Lin and Sheng, Ying and Zhuang, Siyuan and Wu, Zhanghao and Zhuang, Yonghao and Lin, Zi and Li, Zhuohan and Li, Dacheng and Xing, Eric P. and Zhang, Hao and Gonzalez, Joseph E. and Stoica, Ion , year =. Judging. 2306.05685 , archivePrefix=

  126. [150]

    2023 , month = dec, address =

    Liu, Yang and Iter, Dan and Xu, Yichong and Wang, Shuohang and Xu, Ruochen and Zhu, Chenguang , booktitle =. 2023 , month = dec, address =

  127. [151]

    Evaluating Judges as Evaluators: The

    Zhou, Yilun and Xu, Austin and Wang, Peifeng and Xiong, Caiming and Joty, Shafiq , year =. Evaluating Judges as Evaluators: The. 2504.15253 , archivePrefix=

  128. [152]

    PairJudge

    Liu, Yantao and Yao, Zijun and Min, Rui and Cao, Yixin and Hou, Lei and Li, Juanzi , year =. PairJudge. 2501.13007 , archivePrefix=

  129. [153]

    2025 , eprint =

    Inference-Time Reward Hacking in Large Language Models , author =. 2025 , eprint =

  130. [154]

    2025 , eprint =

    Inference-Time Scaling for Generalist Reward Modeling , author =. 2025 , eprint =

  131. [155]

    Proceedings of the 2023 Conference on Empirical Methods in Natural Language Processing , pages =

    Answering Questions by Meta-Reasoning over Multiple Chains of Thought , author =. Proceedings of the 2023 Conference on Empirical Methods in Natural Language Processing , pages =. 2023 , address =

  132. [156]

    2022 , eprint =

    Chain-of-Thought Prompting Elicits Reasoning in Large Language Models , author =. 2022 , eprint =. doi:10.48550/arXiv.2201.11903 , url =

  133. [157]

    2023 , eprint =

    Direct Preference Optimization: Your Language Model is Secretly a Reward Model , author =. 2023 , eprint =. doi:10.48550/arXiv.2305.18290 , url =

  134. [158]

    Advances in Neural Information Processing Systems , year =

    Training Language Models to Follow Instructions with Human Feedback , author =. Advances in Neural Information Processing Systems , year =

  135. [159]

    2023 , eprint =

    Mamba: Linear-Time Sequence Modeling with Selective State Spaces , author =. 2023 , eprint =. doi:10.48550/arXiv.2312.00752 , url =

  136. [161]

    2026 , archivePrefix=

    REALM: Reliable Expertise-Aware Language Model Fine-Tuning from Noisy Annotations , author=. 2026 , archivePrefix=

  137. [162]

    2025 , eprint =

    NVIDIA Nemotron Nano 2: An Accurate and Efficient Hybrid Mamba-Transformer Reasoning Model , author =. 2025 , eprint =. doi:10.48550/arXiv.2508.14444 , url =

  138. [163]

    2025 , eprint =

    EXAONE 4.0: Unified Large Language Models Integrating Non-reasoning and Reasoning Modes , author =. 2025 , eprint =. doi:10.48550/arXiv.2507.11407 , url =

  139. [164]

    deepseek-ai/DeepSeek-R1-Distill-Qwen-32B (Model Card) , howpublished =

  140. [165]

    DeepSeek-R1 (GitHub Repository) , howpublished =

  141. [166]

    Qwen/QwQ-32B (Model Card) , howpublished =

  142. [167]

    2025 , eprint =

    Group Sequence Policy Optimization , author =. 2025 , eprint =. doi:10.48550/arXiv.2507.18071 , url =

  143. [168]

    2025 , eprint =

    Qwen3 Technical Report , author =. 2025 , eprint =. doi:10.48550/arXiv.2505.09388 , url =

  144. [169]

    Qwen/Qwen3-4B-Thinking-2507 (Model Card) , howpublished =

  145. [170]

    Qwen/Qwen3-30B-A3B-Thinking-2507 (Model Card) , howpublished =

  146. [171]

    Qwen/Qwen3-Next-80B-A3B-Thinking (Model Card) , howpublished =

  147. [172]

    gpt-oss-120b & gpt-oss-20b Model Card , howpublished =

  148. [173]

    2025 , eprint =

    gpt-oss-120b & gpt-oss-20b Model Card , author =. 2025 , eprint =. doi:10.48550/arXiv.2508.10925 , url =

  149. [174]

    Introducing gpt-oss , howpublished =

  150. [175]

    openai/gpt-oss-20b (Model Card) , howpublished =

  151. [176]

    gpt-oss-20b Model (OpenAI API Documentation) , howpublished =

  152. [177]

    OpenAI open-weight models (gpt-oss) , howpublished =

  153. [178]

    openai/gpt-oss-safeguard-20b (Model Card) , howpublished =

  154. [179]

    2025 , eprint =

    LLMs Can Easily Learn to Reason from Demonstrations: Structure, not content, is what matters! , author =. 2025 , eprint =. doi:10.48550/arXiv.2502.07374 , url =

  155. [180]

    2024 , eprint =

    SimPO: Simple Preference Optimization with a Reference-Free Reward , author =. 2024 , eprint =. doi:10.48550/arXiv.2405.14734 , url =

  156. [181]

    2025 , eprint =

    Process Reinforcement through Implicit Rewards , author =. 2025 , eprint =. doi:10.48550/arXiv.2502.01456 , url =

  157. [182]

    2024 , eprint =

    Back to Basics: Revisiting REINFORCE Style Optimization for Learning from Human Feedback in LLMs , author =. 2024 , eprint =. doi:10.48550/arXiv.2402.14740 , url =

  158. [183]

    Bespoke-Stratos-17k Dataset Card , howpublished =

  159. [184]

    Bespoke-Stratos-32B Model Card , howpublished =

  160. [185]

    Bespoke-Stratos-7B Model Card , howpublished =

  161. [186]

    Bespoke Curator: Synthetic Data Curation for Post-Training , howpublished =

  162. [187]

    2025 , eprint =

    OpenThoughts: Data Recipes for Reasoning Models , author =. 2025 , eprint =. doi:10.48550/arXiv.2506.04178 , url =

  163. [188]

    OpenThoughts-114K Dataset Card , howpublished =

  164. [189]

    OpenThoughts2-1M Dataset Card , howpublished =

  165. [190]

    OpenThoughts3-1.2M Dataset Card , howpublished =

  166. [191]

    OpenThinker2-32B Model Card , howpublished =

  167. [192]

    OpenThinker3-7B Model Card , howpublished =

  168. [193]

    2024 , eprint =

    Phi-4 Technical Report , author =. 2024 , eprint =. doi:10.48550/arXiv.2412.08905 , url =

  169. [194]

    2025 , eprint =

    Phi-4-Mini-Reasoning: Exploring the Limits of Small Reasoning Language Models in Math , author =. 2025 , eprint =. doi:10.48550/arXiv.2504.21233 , url =

  170. [195]

    microsoft/Phi-4-mini-flash-reasoning (Model Card) , howpublished =

  171. [196]

    2025 , eprint =

    Decoder-Hybrid-Decoder Architecture for Efficient Reasoning with Long Generation , author =. 2025 , eprint =. doi:10.48550/arXiv.2507.06607 , url =

  172. [197]

    Open-R1: A fully open reproduction of DeepSeek-R1 , author =

  173. [198]

    Mixture-of-Thoughts Dataset Card , author =

  174. [199]

    OpenR1-Distill-7B Model Card , author =

  175. [200]

    OpenR1-Math-220k Dataset Card , author =

  176. [201]

    Open R1: Update \#2 , author =

  177. [202]

    CodeForces-CoTs Dataset Card , author =

  178. [203]

    OlympicCoder-7B Model Card , author =

  179. [204]

    OlympicCoder-32B Model Card , author =

  180. [205]

    Open R1: Update \#3 , author =

  181. [206]

    LIMO: Less is More for Reasoning (code and resources) , howpublished =

  182. [207]

    LIMR: Less is More for RL Scaling (code and resources) , howpublished =

  183. [208]

    GAIR/LIMO-v2 (Model Card) , howpublished =

  184. [209]

    GAIR/LIMO-v2 (Dataset Card) , howpublished =

  185. [210]

    LIMOPro: Reasoning Refinement for Efficient and Effective Test-time Scaling , howpublished =

  186. [211]

    FuseAI: Model Fusion for Large Language Models (code and resources) , howpublished =

  187. [212]

    FuseO1-DeepSeekR1-QwQ-SkyT1-32B-Preview (Model Card) , howpublished =

  188. [213]

    FuseO1-DeepSeekR1-QwQ-SkyT1-Flash-32B-Preview (Model Card) , howpublished =

  189. [214]

    mergekit: Tools for merging pretrained large language models , howpublished =

  190. [215]

    qihoo360/Light-R1-32B (Model Card) , howpublished =

  191. [216]

    qihoo360/Light-R1-7B-DS (Model Card) , howpublished =

  192. [217]

    qihoo360/Light-R1-32B-DS (Model Card) , howpublished =

  193. [218]

    qihoo360/Light-R1-14B-DS (Model Card) , howpublished =

  194. [219]

    Journal of the American Statistical Association , volume =

    Probable Inference, the Law of Succession, and Statistical Inference , author =. Journal of the American Statistical Association , volume =. 1927 , doi =

  195. [220]

    Are Your

    Junnan Liu and Hongwei Liu and Linchen Xiao and Ziyi Wang and Kuikun Liu and Songyang Gao and Wenwei Zhang and Songyang Zhang and Kai Chen , booktitle =. Are Your. 2025 , pages =. doi:10.18653/v1/2025.findings-acl.905 , url =

  196. [221]

    International Conference on Learning Representations , year=

    LiveCodeBench: Holistic and Contamination Free Evaluation of Large Language Models for Code , author=. International Conference on Learning Representations , year=

  197. [222]

    Proceedings of the 56th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers) , pages=

    Hierarchical Neural Story Generation , author=. Proceedings of the 56th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers) , pages=

  198. [223]

    International Conference on Learning Representations , year=

    The Curious Case of Neural Text Degeneration , author=. International Conference on Learning Representations , year=

  199. [224]

    Transactions of the Association for Computational Linguistics , volume=

    Locally Typical Sampling , author=. Transactions of the Association for Computational Linguistics , volume=

  200. [225]

    2025 , month = mar, howpublished =

    Sutawika, Lintang and Schoelkopf, Hailey and Gao, Leo and Abbasi, Baber and Biderman, Stella and Tow, Jonathan and others , title =. 2025 , month = mar, howpublished =. doi:10.5281/zenodo.14970487 , url =

  201. [226]

    Advances in Neural Information Processing Systems , volume =

    Balunovic, Mislav and Dekoninck, Jasper and Petrov, Ivo and Jovanovi. Advances in Neural Information Processing Systems , volume =. 2025 , url =

  202. [227]

    Beyond Benchmarks:

    Dekoninck, Jasper and Jovanovi. Beyond Benchmarks:. 2026 , eprint =. doi:10.48550/arXiv.2605.00674 , url =

  203. [228]

    2025 , url =

    Advances in Neural Information Processing Systems , volume =. 2025 , url =

  204. [229]

    2025 , howpublished =

  205. [230]

    2026 , eprint =

    Kwok, Jacky and Li, Shulu and Atreya, Pranav and Liu, Yuejiang and Jiang, Yixing and Finn, Chelsea and Pavone, Marco and Stoica, Ion and Mirhoseini, Azalia , title =. 2026 , eprint =. doi:10.48550/arXiv.2607.05391 , url =

  206. [231]

    2026 , month = jun, howpublished =

  207. [232]

    2024 , howpublished =

    Jia, Maxwell , title =. 2024 , howpublished =

  208. [233]

    doi:10.57967/hf/2439 , url =

    2024 , howpublished =. doi:10.57967/hf/2439 , url =

  209. [234]

    2026 , howpublished =

  210. [235]

    Phi-4-reasoning technical report

    Marah Abdin, Sahaj Agarwal, Ahmed Awadallah, Vidhisha Balachandran, Harkirat Behl, Lingjiao Chen, Gustavo de Rosa, Suriya Gunasekar, Mojan Javaheripi, Neel Joshi, et al. Phi-4-reasoning technical report. arXiv preprint arXiv:2504.21318, 2025. doi:10.48550/arXiv.2504.21318. URL...

  211. [236]

    Phi-4 technical report, 2024

    Marah Abdin et al. Phi-4 technical report, 2024. URL https://arxiv.org/abs/2412.08905

  212. [237]

    Let ' s sample step by step: Adaptive-consistency for efficient reasoning and coding with LLM s

    Pranjal Aggarwal, Aman Madaan, Yiming Yang, and Mausam. Let ' s sample step by step: Adaptive-consistency for efficient reasoning and coding with LLM s. In Proceedings of the 2023 Conference on Empirical Methods in Natural Language Processing, pp.\ 12375--12396, Singapore, Dec...

  213. [238]

    OpenCodeReasoning-II: A Simple Test Time Scaling Approach via Self-Critique , 2025 a

    Wasi Uddin Ahmad, Somshubra Majumdar, Aleksander Ficek, Sean Narenthiran, Mehrzad Samadi, Jocelyn Huang, Siddhartha Jain, Vahid Noroozi, and Boris Ginsburg. OpenCodeReasoning-II: A Simple Test Time Scaling Approach via Self-Critique , 2025 a . URL https://arxiv.org/abs/2507.09075

  214. [239]

    OpenCodeReasoning: Advancing Data Distillation for Competitive Coding

    Wasi Uddin Ahmad, Sean Narenthiran, Somshubra Majumdar, Aleksander Ficek, Siddhartha Jain, Jocelyn Huang, Vahid Noroozi, and Boris Ginsburg. OpenCodeReasoning: Advancing Data Distillation for Competitive Coding . arXiv preprint arXiv:2504.01943, 2025 b . URL https://arxiv.org/...

  215. [240]

    Back to basics: Revisiting reinforce style optimization for learning from human feedback in llms, 2024

    Arash Ahmadian, Chris Cremer, Matthias Gall \'e , Marzieh Fadaee, Julia Kreutzer, Olivier Pietquin, Ahmet \"U st \"u n, and Sara Hooker. Back to basics: Revisiting reinforce style optimization for learning from human feedback in llms, 2024. URL https://arxiv.org/abs/2402.14740

  216. [241]

    mergekit: Tools for merging pretrained large language models

    Arcee AI . mergekit: Tools for merging pretrained large language models. https://github.com/arcee-ai/mergekit, 2024. Accessed: 2026-02-26

  217. [242]

    MathArena : Evaluating llms on uncontaminated math competitions

    Mislav Balunovic, Jasper Dekoninck, Ivo Petrov, Nikola Jovanovi \'c , and Martin Vechev. MathArena : Evaluating llms on uncontaminated math competitions. In Advances in Neural Information Processing Systems, volume 38, 2025. URL https://proceedings.neurips.cc/paper_files/paper...

  218. [243]

    Nvidia nemotron nano 2: An accurate and efficient hybrid mamba-transformer reasoning model

    Aarti Basant, Abhijit Khairnar, Abhijit Paithankar, Abhinav Khattar, Adithya Renduchintala, Aditya Malte, Akhiad Bercovich, Akshay Hazare, Alejandra Rico, Aleksander Ficek, et al. Nvidia nemotron nano 2: An accurate and efficient hybrid mamba-transformer reasoning model. arXiv...

  219. [244]

    Amanda Bertsch, Alex Xie, Graham Neubig, and Matthew R. Gormley. It's MBR all the way down: Modern generation techniques through the lens of minimum bayes risk, 2023. URL https://arxiv.org/abs/2310.01387

  220. [245]

    Bespoke curator: Synthetic data curation for post-training

    Bespoke Labs . Bespoke curator: Synthetic data curation for post-training. https://github.com/bespokelabsai/curator, 2025 a . Accessed: 2025-01-31

  221. [246]

    Bespoke-stratos: The unreasonable effectiveness of reasoning distillation

    Bespoke Labs . Bespoke-stratos: The unreasonable effectiveness of reasoning distillation. https://www.bespokelabs.ai/blog/bespoke-stratos-the-unreasonable-effectiveness-of-reasoning-distillation, 2025 b . Accessed: 2025-01-22

  222. [247]

    Bespoke-stratos-17k dataset card

    Bespoke Labs . Bespoke-stratos-17k dataset card. https://huggingface.co/datasets/bespokelabs/Bespoke-Stratos-17k, 2025 c . Accessed: 2025-01-31

  223. [248]

    Bespoke-stratos-32b model card

    Bespoke Labs . Bespoke-stratos-32b model card. https://huggingface.co/bespokelabs/Bespoke-Stratos-32B, 2025 d . Accessed: 2025-01-31

  224. [249]

    Bespoke-stratos-7b model card

    Bespoke Labs . Bespoke-stratos-7b model card. https://huggingface.co/bespokelabs/Bespoke-Stratos-7B, 2025 e . Accessed: 2025-01-31

  225. [250]

    We need to talk about random seeds

    Steven Bethard. We need to talk about random seeds. arXiv preprint arXiv:2210.13393, 2022. doi:10.48550/arXiv.2210.13393. URL https://arxiv.org/abs/2210.13393

  226. [251]

    Towards reproducible llm evaluation: Quantifying uncertainty in llm benchmark scores

    Robert E Blackwell, Jon Barry, and Anthony G Cohn. Towards reproducible llm evaluation: Quantifying uncertainty in llm benchmark scores. arXiv preprint arXiv:2410.03492, 2024

  227. [252]

    Accounting for variance in machine learning benchmarks

    Xavier Bouthillier, Pierre Delaunay, Mirko Bronzi, Assya Trofimov, Brennan Nichyporuk, Justin Szeto, Nazanin Mohammadi Sepahvand, Edward Raff, Kanika Madan, Vikram Voleti, Samira Ebrahimi Kahou, Vincent Michalski, Dmitriy Serdyuk, Tal Arbel, Chris Pal, Ga \"e l Varoquaux, and ...

  228. [253]

    Browne, Edward Powley, Daniel Whitehouse, Simon M

    Cameron B. Browne, Edward Powley, Daniel Whitehouse, Simon M. Lucas, Peter I. Cowling, Philipp Rohlfshagen, Stephen Tavener, Diego Perez, Spyridon Samothrakis, and Simon Colton. A survey of M onte C arlo tree search methods. IEEE Transactions on Computational Intelligence and ...

  229. [254]

    Evaluating large language models trained on code

    Mark Chen, Jerry Tworek, Heewoo Jun, Qiming Yuan, Henrique Ponde De Oliveira Pinto, Jared Kaplan, Harri Edwards, Yuri Burda, Nicholas Joseph, Greg Brockman, et al. Evaluating large language models trained on code. arXiv preprint arXiv:2107.03374, 2021

  230. [255]

    Universal self-consistency for large language model generation, 2023

    Xinyun Chen, Renat Aksitov, Uri Alon, Jie Ren, Kefan Xiao, Pengcheng Yin, Sushant Prakash, Charles Sutton, Xuezhi Wang, and Denny Zhou. Universal self-consistency for large language model generation, 2023. URL https://arxiv.org/abs/2311.17311

  231. [256]

    Faster minimum bayes risk decoding with confidence-based pruning, 2023

    Julius Cheng and Andreas Vlachos. Faster minimum bayes risk decoding with confidence-based pruning, 2023. URL https://arxiv.org/abs/2311.14919

  232. [257]

    Chatbot arena: An open platform for evaluating llms by human preference

    Wei-Lin Chiang, Lianmin Zheng, Ying Sheng, Anastasios Nikolas Angelopoulos, Tianle Li, Dacheng Li, Banghua Zhu, Hao Zhang, Michael Jordan, Joseph E Gonzalez, et al. Chatbot arena: An open platform for evaluating llms by human preference. In Forty-first International Conference...

  233. [258]

    Training verifiers to solve math word problems, 2021

    Karl Cobbe et al. Training verifiers to solve math word problems, 2021. URL https://arxiv.org/abs/2110.14168

  234. [259]

    Process reinforcement through implicit rewards, 2025

    Ganqu Cui, Lifan Yuan, Zefan Wang, Hanbin Wang, Yuchen Zhang, Jiacheng Chen, Wendi Li, Bingxiang He, Yuchen Fan, Tianyu Yu, Qixin Xu, Weize Chen, Jiarui Yuan, Huayu Yuan, Kaiyan Zhang, Xingtai Lv, Shuo Wang, Yuan Yao, Xu Han, Hao Peng, Yu Cheng, Zhiyuan Liu, Maosong Sun, Bowen...

  235. [260]

    deepseek-ai/deepseek-r1-distill-qwen-32b (model card)

    DeepSeek-AI . deepseek-ai/deepseek-r1-distill-qwen-32b (model card). https://huggingface.co/deepseek-ai/DeepSeek-R1-Distill-Qwen-32B, 2025 a . Accessed: 2026-02-26

  236. [261]

    Deepseek-r1 (github repository)

    DeepSeek-AI . Deepseek-r1 (github repository). https://github.com/deepseek-ai/DeepSeek-R1, 2025 b . Accessed: 2026-02-26

  237. [262]

    Beyond benchmarks: MathArena as an evaluation platform for mathematics with llms, 2026

    Jasper Dekoninck, Nikola Jovanovi \'c , Tim Gehrunger, K \'a ri R \"o gnvaldsson, Ivo Petrov, Chenhao Sun, and Martin Vechev. Beyond benchmarks: MathArena as an evaluation platform for mathematics with llms, 2026. URL https://arxiv.org/abs/2605.00674

  238. [263]

    Best-of-majority: Minimax-optimal strategy for pass@ k inference scaling

    Qiwei Di, Kaixuan Ji, Xuheng Li, Heyang Zhao, and Quanquan Gu. Best-of-majority: Minimax-optimal strategy for pass@ k inference scaling. In The Fourteenth International Conference on Learning Representations, 2026. URL https://openreview.net/forum?id=a6CVQpjbXq

  239. [264]

    Jesse Dodge, Suchin Gururangan, Dallas Card, Roy Schwartz, and Noah A. Smith. Show your work: Improved reporting of experimental results. In Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing and the 9th International Joint Conference on Nat...

  240. [265]

    Bootstrap methods: Another look at the jackknife

    Bradley Efron. Bootstrap methods: Another look at the jackknife. The Annals of Statistics, 7 0 (1): 0 1--26, 1979

  241. [266]

    Is MAP decoding all you need? the inadequacy of the mode in neural machine translation

    Bryan Eikema and Wilker Aziz. Is MAP decoding all you need? the inadequacy of the mode in neural machine translation. In Proceedings of the 28th International Conference on Computational Linguistics (COLING), 2020. URL https://aclanthology.org/2020.coling-main.398/

  242. [267]

    Hierarchical neural story generation

    Angela Fan, Mike Lewis, and Yann Dauphin. Hierarchical neural story generation. In Proceedings of the 56th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), pp.\ 889--898, 2018

  243. [268]

    High quality rather than high model probability: Minimum bayes risk decoding with neural metrics

    Markus Freitag, David Grangier, Qijun Tan, and Bowen Liang. High quality rather than high model probability: Minimum bayes risk decoding with neural metrics. Transactions of the Association for Computational Linguistics, 10: 0 811--825, 2022. doi:10.1162/tacl_a_00491. URL http...

  244. [269]

    Deep think with confidence

    Yichao Fu, Xuewei Wang, Hao Zhang, Yuandong Tian, and Jiawei Zhao. Deep think with confidence. In The Fourteenth International Conference on Learning Representations, 2026. URL https://openreview.net/forum?id=8LqHs0KIM7

  245. [270]

    Fuseo1-deepseekr1-qwq-skyt1-32b-preview (model card)

    FuseAI . Fuseo1-deepseekr1-qwq-skyt1-32b-preview (model card). https://huggingface.co/FuseAI/FuseO1-DeepSeekR1-QwQ-SkyT1-32B-Preview, 2025 a . Accessed: 2026-02-26

  246. [271]

    Fuseo1-deepseekr1-qwq-skyt1-flash-32b-preview (model card)

    FuseAI . Fuseo1-deepseekr1-qwq-skyt1-flash-32b-preview (model card). https://huggingface.co/FuseAI/FuseO1-DeepSeekR1-QwQ-SkyT1-Flash-32B-Preview, 2025 b . Accessed: 2026-02-26

  247. [272]

    Fuseai: Model fusion for large language models (code and resources)

    FuseAI . Fuseai: Model fusion for large language models (code and resources). https://github.com/fanqiwan/FuseAI, 2025 c . Accessed: 2026-02-26

  248. [273]

    Gair/limo-v2 (dataset card)

    GAIR . Gair/limo-v2 (dataset card). https://huggingface.co/datasets/GAIR/LIMO-v2, 2025 a . Accessed: 2026-02-26

  249. [274]

    Gair/limo-v2 (model card)

    GAIR . Gair/limo-v2 (model card). https://huggingface.co/GAIR/LIMO-v2, 2025 b . Accessed: 2026-02-26

  250. [275]

    Scaling laws for reward model overoptimization

    Leo Gao, John Schulman, and Jacob Hilton. Scaling laws for reward model overoptimization. In Proceedings of the 40th International Conference on Machine Learning (ICML), volume 202 of Proceedings of Machine Learning Research, pp.\ 10835--10866, 2023. URL https://proceedings.ml...

  251. [276]

    Decentralized low-rank fine-tuning of large language models

    Sajjad Ghiasvand, Mahnoosh Alizadeh, and Ramtin Pedarsani. Decentralized low-rank fine-tuning of large language models. In Proceedings of the 1st Workshop for Research on Agent Language Models (REALM 2025). Association for Computational Linguistics, July 2025. doi:10.18653/v1/...

  252. [277]

    Realm: Reliable expertise-aware language model fine-tuning from noisy annotations, 2026

    Sajjad Ghiasvand, Mark Beliaev, Mahnoosh Alizadeh, and Ramtin Pedarsani. Realm: Reliable expertise-aware language model fine-tuning from noisy annotations, 2026. URL https://arxiv.org/abs/2604.17289

  253. [278]

    Evalchemy , November 2024

    Etash Guha, Negin Raoof, Jean Mercat, Eric Frankel, Sedrick Keh, Sachin Grover, George Smyrnis, Trung Vu, Ryan Marten, Jon Saad-Falcon, Caroline Choi, Kushal Arora, Mike Merrill, Yichuan Deng, Ashima Suvarna, Hritik Bansal, Marianna Nezhurina, Yejin Choi, Reinhard Heckel, Seew...

  254. [279]

    Openthoughts: Data recipes for reasoning models, 2025

    Etash Guha, Ryan Marten, Sedrick Keh, Negin Raoof, Georgios Smyrnis, Hritik Bansal, Marianna Nezhurina, Jean Mercat, Trung Vu, Zayne Sprague, et al. Openthoughts: Data recipes for reasoning models, 2025. URL https://arxiv.org/abs/2506.04178

  255. [280]

    Deepseek-r1: Incentivizing reasoning capability in llms via reinforcement learning

    Daya Guo, Dejian Yang, Haowei Zhang, Junxiao Song, Ruoyu Zhang, Runxin Xu, Qihao Zhu, Shirong Ma, Peiyi Wang, Xiao Bi, et al. Deepseek-r1: Incentivizing reasoning capability in llms via reinforcement learning. arXiv preprint arXiv:2501.12948, 2025

  256. [281]

    Reasoning with language model is planning with world model

    Shibo Hao, Yi Gu, Haodi Ma, Joshua Hong, Zhen Wang, Daisy Zhe Wang, and Zhiting Hu. Reasoning with language model is planning with world model. In Proceedings of the 2023 Conference on Empirical Methods in Natural Language Processing, pp.\ 8154--8173, Singapore, December 2023....

  257. [282]

    Ranking reasoning LLM s under test-time scaling

    Mohsen Hariri, Michael Hinczewski, Jing Ma, and Vipin Chaudhary. Ranking reasoning LLM s under test-time scaling. In Maria Liakata, Viviane P. Moreira, Jiajun Zhang, and David Jurgens (eds.), Proceedings of the 64th Annual Meeting of the A ssociation for C omputational L ingui...

  258. [283]

    Quantize what counts: More for keys, less for values

    Mohsen Hariri, Alan Luo, Weicong Chen, Tianyi Zhang, Qifan Wang, Xiaotian Han, and Vipin Chaudhary. Quantize what counts: More for keys, less for values. In Maria Liakata, Viviane P. Moreira, Jiajun Zhang, and David Jurgens (eds.), Findings of the A ssociation for C omputation...

  259. [284]

    Don't pass@k: A bayesian framework for large language model evaluation

    Mohsen Hariri, Amirhossein Samandar, Michael Hinczewski, and Vipin Chaudhary. Don't pass@k: A bayesian framework for large language model evaluation. In Proceedings of the 14th International Conference on Learning Representations (ICLR 2026), 2026 c . URL https://openreview.ne...

  260. [285]

    Truncation sampling as language model desmoothing

    John Hewitt, Christopher D Manning, and Percy Liang. Truncation sampling as language model desmoothing. arXiv preprint arXiv:2210.15191, 2022

  261. [286]

    A sober look at progress in language model reasoning: Pitfalls and paths to reproducibility

    Andreas Hochlehnert, Hardik Bhatnagar, Vishaal Udandarao, Samuel Albanie, Ameya Prabhu, and Matthias Bethge. A sober look at progress in language model reasoning: Pitfalls and paths to reproducibility. In Second Conference on Language Modeling, 2025. URL https://openreview.net...

  262. [287]

    The curious case of neural text degeneration

    Ari Holtzman, Jan Buys, Li Du, Maxwell Forbes, and Yejin Choi. The curious case of neural text degeneration. In International Conference on Learning Representations, 2020

  263. [288]

    Is best-of- n the best of them? coverage, scaling, and optimality in inference-time alignment

    Audrey Huang, Adam Block, Qinghua Liu, Nan Jiang, Akshay Krishnamurthy, and Dylan Foster. Is best-of- n the best of them? coverage, scaling, and optimality in inference-time alignment. In Proceedings of the 42nd International Conference on Machine Learning (ICML), volume 267 o...

  264. [289]

    Open r1: A fully open reproduction of deepseek-r1, January 2025 a

    Hugging Face . Open r1: A fully open reproduction of deepseek-r1, January 2025 a . URL https://github.com/huggingface/open-r1

  265. [290]

    Open-r1: A fully open reproduction of deepseek-r1

    Hugging Face . Open-r1: A fully open reproduction of deepseek-r1. https://github.com/huggingface/open-r1, 2025 b . Accessed: 2026-02-26

  266. [291]

    Open r1: Update \#2

    Hugging Face . Open r1: Update \#2. https://huggingface.co/blog/open-r1/update-2, 2025 c . Accessed: 2026-02-26

  267. [292]

    Open r1: Update \#3

    Hugging Face . Open r1: Update \#3. https://huggingface.co/blog/open-r1/update-3, 2025 d . Accessed: 2026-02-26

  268. [293]

    Reward-guided tree search for inference time alignment of large language models

    Chia-Yu Hung, Navonil Majumder, Ambuj Mehrish, and Soujanya Poria. Reward-guided tree search for inference time alignment of large language models. In Proceedings of the 2025 Conference of the North American Chapter of the Association for Computational Linguistics: Human Langu...

  269. [294]

    Wang, Armando Solar-Lezama, Koushik Sen, and Ion Stoica

    Naman Jain, King Han, Alex Gu, Wen-Ding Li, Fanjia Yan, Tianjun Zhang, Sida I. Wang, Armando Solar-Lezama, Koushik Sen, and Ion Stoica. Livecodebench: Holistic and contamination free evaluation of large language models for code. In International Conference on Learning Represen...

  270. [295]

    AIME 2024 dataset

    Maxwell Jia. AIME 2024 dataset. Hugging Face dataset, 2024. URL https://huggingface.co/datasets/Maxwell-Jia/AIME_2024

  271. [296]

    Hyperparameter-free approach for faster minimum bayes risk decoding, 2024

    Yuu Jinnai and Kaito Ariu. Hyperparameter-free approach for faster minimum bayes risk decoding, 2024. URL https://arxiv.org/abs/2401.02749

  272. [297]

    Regularized best-of- n sampling with minimum B ayes risk objective for language model alignment

    Yuu Jinnai, Tetsuro Morimura, Kaito Ariu, and Kenshi Abe. Regularized best-of- n sampling with minimum B ayes risk objective for language model alignment. In Proceedings of the 2025 Conference of the Nations of the Americas Chapter of the Association for Computational Linguist...

  273. [298]

    Scalable best-of- n selection for large language models via self-certainty

    Zhewei Kang, Xuandong Zhao, and Dawn Song. Scalable best-of- n selection for large language models via self-certainty. In Advances in Neural Information Processing Systems, volume 38, 2025. URL https://proceedings.neurips.cc/paper_files/paper/2025/hash/1c7eff166a8e345f664f0faa...

  274. [299]

    Inference-time reward hacking in large language models, 2025

    Hadi Khalaf, Claudio Mayrink Verdun, Alex Oesterling, Himabindu Lakkaraju, and Flavio du Pin Calmon. Inference-time reward hacking in large language models, 2025. URL https://arxiv.org/abs/2506.19248. Accepted to NeurIPS 2025 (Spotlight)

  275. [300]

    Simulating a bias mitigation scenario in large language models, 2025

    Kiana Kiashemshaki, Mohammad Jalili Torkamani, Negin Mahmoudi, and Meysam Shirdel Bilehsavar. Simulating a bias mitigation scenario in large language models, 2025. URL https://arxiv.org/abs/2509.14438

Pith tools

Reviewed August 15, 2026 · model on record in the stance chip above.